# From Stress-Relief Vibration Patterns to Haptic Emojis, USC Researchers Personalize the Sense of Touch With AI > Source: > Published: 2026-07-27 18:22:25+00:00 From the vibration that warns you when your car drifts out of its lane to the buzz of your iPhone alarm or the rumble of a video game controller that makes virtual worlds feel real, haptic vibration plays an important role in everyday life. Haptic feedback refers to technology that communicates through the sense of touch. Much like music can evoke emotions through different melodies and rhythms, humans naturally associate different vibration patterns with specific meanings and emotional responses. Different combinations of vibration frequency, amplitude and waveform can create entirely different tactile sensations. While haptic feedback is embedded in technologies ranging from smartphones and wearables to gaming systems and virtual reality, the same vibration pattern can feel very different from one person to another. A vibration designed as a gentle notification may be barely noticeable to one user but feel distracting or anxiety-inducing to another. These individual differences often make predefined haptic libraries ineffective at delivering the intended experience. Although personalized haptic feedback could create more intuitive and accessible user experiences, efficiently identifying each person’s preferred vibration patterns has remained a longstanding challenge. USC haptics researcher [Heather Culbertson](https://viterbi.usc.edu/directory/faculty/Culbertson/Heather) aimed to close this gap with her latest study. Led by Culbertson, the research team developed an algorithm called Vibrotactile Preference Learning (VPL), a system that efficiently learns an individual’s preferred vibration patterns for producing a desired emotional response. The yearlong project resulted in the paper, “[Vibrotactile Preference Learning: Uncertainty-Aware Preference Learning for Personalized Vibration Feedback](https://dl.acm.org/doi/full/10.1145/3774935.3806784),” which was recently accepted to the [ACM Conference on User Modeling, Adaptation and Personalization](https://www.um.org/umap2026/) 2026. Culbertson is an associate professor of computer science with joint appointments in [USC Viterbi School of Engineering](https://viterbi.usc.edu/) and the [USC Mark and Mary Stevens School of Computing and AI](https://stevens-computing-ai.usc.edu/)‘s [Thomas Lord Department of Computer Science](https://www.cs.usc.edu/), the [Alfred E. Mann Department of Biomedical Engineering](https://bme.usc.edu/) and [department of Aerospace and Mechanical Engineering](https://ame.usc.edu/). Culbertson also directs the Haptics, Robotics, and Virtual Interaction [(HaRVI) Lab](https://sites.usc.edu/culbertson/) at USC. The study was conducted in collaboration with [Erdem Bıyık](https://viterbi.usc.edu/directory/faculty/Biyik/Erdem), USC assistant professor of computer science and electrical and computer engineering. ## Why Personalizing Haptic Feedback Has Been So Difficult Individual differences in physiology, including factors such as skin thickness and the density of touch receptors, as well as differences in psychology can significantly influence how people perceive the same haptic signal. This phenomenon is called tactile subjectivity. To account for these differences, many previous studies have relied on traditional “method of adjustment” interfaces. In these systems, participants manually adjust technical parameters using sliders until they reach their preferred haptic sensation. However, this approach is often unintuitive and struggles to accurately capture user preferences. Most users are unfamiliar with technical parameters such as frequency and amplitude, making it difficult for them to navigate the large number of possible vibration settings. Other conventional approaches ask users to rate vibrations on an absolute scale, such as a 1-to-7 Likert scale. These methods can suffer from what researchers call score drift, where users gradually lose a consistent internal reference after making many ratings, leading to fatigue and less reliable responses. Previous research has also shown that people are generally better at making pairwise comparisons, similar to choosing between option A or option B, than assigning numerical ratings. However, determining which pairs of vibration signals users should compare while minimizing the total number of comparisons has remained a significant challenge for researchers. ## AI Algorithm Makes Personalized Haptics Faster and More Efficient To address this longstanding challenge, Culbertson’s team developed a machine learning approach that dramatically reduces the number of vibration comparisons users need to make while maximizing the information gained from each response, helping minimize user fatigue. The researchers created an algorithm called Vibrotactile Preference Learning (VPL), which uses an active querying strategy to identify a user’s preferred vibration pattern in as few as 40 rounds of pairwise comparisons. By comparison, traditional methods may require nearly 200 comparisons to identify a favorite among just 20 vibration signals. The system first represents each vibration using four key parameters in what researchers call a four-dimensional space: intensity, motor balance (texture), rhythm (pulse frequency) and grain (pulse duty cycle). VPL then uses a Gaussian process preference model to estimate a user’s preferences across that four-dimensional space. Rather than asking users to compare every possible vibration pair, the algorithm intelligently selects the next pair expected to provide the most useful information. After each comparison, users also report how confident they are in their choice using a five-point scale. The algorithm incorporates those confidence scores when updating its predictions, placing greater weight on responses made with higher confidence while accounting for more uncertain comparisons. By continually selecting the most informative comparisons, the system efficiently searches the vast design space to identify personalized vibration patterns with far fewer user interactions. Rather than identifying a single “best” vibration, the VPL system can recommend personalized haptic feedback tailored to a specific emotion, sensation or design goal defined by the haptic designer. ## From Stress-Relief Vibrations to Haptic Emojis: Personalized Haptics in the Real World The algorithm developed by Culbertson’s team could be used to design and personalize haptic feedback across a wide range of applications, from immersive gaming and mental health to automotive safety and digital communication. In mental health and wellness, the technology could personalize vibrations for meditation, therapy and stress-management apps, creating tactile feedback tailored to help users feel calmer, regulate stress or support guided breathing exercises. In gaming and virtual or augmented reality, VPL could customize vibrations to realistically simulate events such as collisions, explosions or driving over rough terrain. Developers could also design distinct haptic patterns for different in-game actions while tailoring vibrations to evoke emotions such as excitement, fear or calmness based on users’ preferences. When it comes to road safety, the system could personalize vibration-based alerts for lane departures, collision warnings and navigation cues, ensuring critical notifications are both noticeable and appropriately matched to their level of urgency without becoming unnecessarily intrusive. The research could also enable more personalized forms of digital communication through “haptic emojis” — customized vibration patterns that convey emotions such as affection, comfort or excitement. The technology could even create unique haptic signatures for different contacts, allowing users to recognize whether a call is from a partner, parent or close friend without looking at a screen. Published on July 27th, 2026 Last updated on July 27th, 2026